This is a vote dilution case. African American voters in Glades County, Florida, challenge the at-large method of electing members of the County Commission and School Board, claiming that it depreciates their right to vote on account of their race in violation of § 2 of the Voting Rights Act of 1965, 42 U.S.C. § 1973, and the Fourteenth and Fifteenth Amendments. Following a bench trial, the United States District Court for the Middle District of Florida denied relief. The Plaintiffs now appeal the court’s judgment. We reverse and remand.
I.
A.
Glades County spans approximately 988.2 square miles of southwest Florida. According to the 2000 decennial census, the county had an overall population of 10,576, and a voting age population of 8,329. Of the overall population, 69.4% were white, 15.0% were Hispanic, and 10.5% were African American. Of the voting age population, 73.2% were white, 12.7% were Hispanic, and 10.1% were African American.1 Population numbers alone, however, do not portray the unique geographic character of the minority population in Glades County. Most of the Hispanic population is dispersed across the southern portion of the county, but the African American population is concentrated in an area known as Washington Park, which sits in Moore Haven, the county’s seat and largest city.
The County Commission and the County School Board both consist of five elected members who serve staggered, four-year terms. Likewise, the same election scheme governs both the County Commission and the School Board: the county is divided into five residential districts, and candidates run from the district in which they live, but at-large voting determines the outcome of each election.2 Candidates that receive a majority of the countywide vote in a primary election are selected as their political party’s nominee,3 and a plurality of the countywide vote is sufficient to win the general election.
During the 1998 Democratic Primary for School Board District Four, Mike Pressley won the two-candidate contest with 58.2% of the vote, defeating plaintiff Billie Thompson, who garnered 41.8% of the vote.4 Prior to 1998, however, the only other African American candidate for *1257countywide office on either the County Commission or School Board was Charles Hall, who won a seat (by a nine-vote margin during a run-off primary election) on the County Commission in 1976.
B.
This law suit commenced on May 4, 2000. The Plaintiffs, Billie Thompson and Patricia Brown, African Americans, reside in Glades County and are registered to vote there. The defendants consist of the Board of County Commissioners and its five members, the County School Board and its five members, and the Supervisor of Elections. The complaint contained three “causes of action.”5 The first is that the at-large schemes for the election of members of the County Commission and School Board were adopted by the State of Florida and have been maintained for the purpose of diluting, minimizing, and canceling out the voting strength of African Americans in violation of § 2 of the Voting Rights Act.6 The second and third causes of action allege, respectively, that the same at-large electoral schemes deny the Plaintiffs the equal protection of the laws guaranteed by the Fourteenth Amendment,7 and their rights under the Fifteenth Amendment.8 For relief, the complaint sought a declaratory judgment that the challenged schemes were invalid under § 2 and the Fourteenth and Fifteenth Amendments, injunction against their further use, and the creation of “new procedures or plans for election” of the members of the County Commission and School Board.
The defendants, in their answers, denied that the challenged electoral schemes diluted the Plaintiffs’ right to vote, and asserted, moreover, that a remedy could not lawfully be created to enhance the Plaintiffs’ voting power. As the case proceeded toward trial, the Plaintiffs proposed what they contended would be a remedy that met the requirements of law, specifically the Supreme Court’s decision in Thornburg v. Gingles, 478 U.S. 30, 106 S.Ct. 2752, 92 L.Ed.2d 25 (1986): the County Commissioners and School Board members would reside in and be elected from five single-member districts, one of which would contain an African American majority among the voting age population.
The filing of this law suit invited a nonbinding special referendum that was held on June 5, 2001. Two questions were presented: (1) “Shall the five members of the Board of Commissioners of Glades County, Florida, be elected to office from single-member districts by the electors residing in each of those districts only?”; and (2) “Shall the five members of the School Board of Glades County, Florida, be elected to office from single-member districts *1258by electors residing in each of those districts only?” A majority of those voting opposed single-member districts. The final tally for the County Commission indicated 825 votes (53.4%) opposed to single-member districts, and 719 votes (46.6%) in favor of changing the election scheme. Of these votes, in precincts with an African American voting age population of 90% or greater, 52% voted in favor of single-member districts; in precincts with a white voting age population of 90% or greater, 48% voted in favor of single-member districts. With regard to the School Board, the final tally indicated 848 votes (54.9%) opposed to single-member districts, and 698 votes (45.1%) in favor of changing the election scheme. Similarly, in precincts with an African American voting age population of 90% or greater, 52% voted in favor of single-member districts; in precincts with a white voting age population of 90% or greater, 47% voted in favor of single-member districts.
The case went to trial before the district court on October 10, 2001. The court heard the testimony of four witnesses (the Plaintiffs and two of their experts), received the Plaintiffs’ exhibits, which included the reports of a third expert, and the parties’ stipulations, and then recessed the proceeding to await the filing of depositions. Over the next several days, the Plaintiffs filed the depositions of eight members of the County Commission and School Board and the Supervisor of Elections. On November 14, the court reconvened the proceeding to hear the parties’ closing arguments, which focused on the Plaintiffs’ proposed illustrative redistricting plan and the defendants’ reasons why the plan could not be lawfully implemented.
The Plaintiffs relied on three expert witnesses: (1) William S. Cooper, who prepared the Plaintiffs’ illustrative plan9 for five single-member districts;10 (2) Dr. Steven P. Cole, who analyzed statistics relating to Glades County voter behavior;11 and (3) Dr. Gary R. Mormino, who provided socio-political background to Florida’s adoption of statutes and constitutional provisions affecting the method of election to the County Commission and School Board.
The illustrative plan proposed by the Plaintiffs contains five new single-member districts. Among the new districts, District One contains a slim African American voting age majority. Specifically, 50.23% of the voting age population in District One is African American12 and 15.23% is Hispanic. The total minority voting age population in District One (inclusive of oth*1259er minorities such as American Indians) is approximately 67%. With a total population of 10,576 in Glades County, each of the five new districts would ideally have a population approximating 2,115 in order to comport with the one-person, one-vote principle under the Fourteenth Amendment. Among the five districts, the plan has an overall deviation of 8.6%, with District One underpopulated by 86 individuals. The following table summarizes the salient features of the Plaintiffs’ illustrative plan:
District Total Population Deviation (persons)Percent Voting Age Hispanics Percent Voting Age African Americans
1_2,029 -86_15.23% 50.23%
2_2¿83_+68 13.30% 1.95%
3_2211_+96_1.81% 0.44%
4_2,080 -35_1.99% 0.23%
5 2,073 -42_35.19% 0.95%
The distinctive feature of this illustrative plan is that nearly the entire 834-person population of voting age African Americans in Glades County is contained within District One, which is, itself, dwarfed in size by the other districts (due to the large concentration of African Americans in Moore Haven). In sum, 775 of the 834 voting age African Americans (92.9%) would be residents of the proposed District One.
C.
The district court issued its decision in this case on August 27, 2004, in the form of
a 49-page Opinion and Order. The court concluded that the Plaintiffs failed to make out a case of vote dilution under § 2 of the Voting Rights Act or violations of the Fourteenth and Fifteenth Amendments.13 With respect to their § 2 claim, the court found (1) that District One constituted an “influence district” impermissible as a remedy under § 2 of the Act, and (2) that the illustrative plan did not constitute a permissible § 2 remedy because its 8.6% overall deviation did not comport with the one-person, one-vote requirement of the Fourteenth Amendment.
The Clerk of the District Court entered judgment pursuant to the court’s Opinion and Order on August 30, 2004. The Plaintiffs thereafter moved the court to alter or amend its judgment pursuant to Federal Rule of Civil Procedure 59(e), questioning the court’s findings regarding their § 2 remedy. They emphasized a point the court had not addressed: the possibility that cross-over votes by whites and Hispanics could enable African Americans to elect their candidates of choice. In their motion, the Plaintiffs cited a regression analysis that accounted for the Hispanic vote,14 in which Dr. Cole concluded that African American and Hispanic voters are politically cohesive. According to that analysis, an estimated 74% of African American and Hispanic voters combined voted for the African American candidate *1260in the 1998 Democratic Primary for School Board District Four, in contrast with the 15% of non-Hispanic, non-African American voters who voted for the African American candidate.
The district court denied the Plaintiffs’ Rule 59(e) motion, concluding that the Plaintiffs failed to establish any clear error in the court’s decision, or that manifest injustice would result. This appeal followed.
II.
In this appeal, the Plaintiffs do not challenge the district court’s disposition of their constitutional claims. Instead, they limit their appeal to the court’s denial of their claim under § 2 of the Voting Rights Act. In relevant part, the Act declares:
(a) No voting qualification or prerequisite to voting or standard, practice, or procedure shall be imposed or applied by any State or political subdivision in a manner which results in a denial or abridgement of the right of any citizen of the United States to vote on account of race or color, or in contravention of the guarantees set forth in section 1973b(f)(2) of this title, as provided in subsection (b) of this section.
(b) A violation of subsection (a) of this section is established if, based on the totality of circumstances, it is shown that the political processes leading to ... election in the State ... are not equally open to participation by members of a class of citizens protected by subsection (a) of this section in that its members have less opportunity than other members of the electorate to participate in the political process and to elect representatives of their choice. The extent to which members of a protected class have been elected to office in the State or political subdivision is one circumstance which may be considered: Provided, That nothing in this section establishes a right to have members of a protected class elected in numbers equal to their proportion in the population.
42 U.S.C. § 1973.
Under § 2, vote dilution occurs where “an election practice results in the dilution of minority voting strength and, thus, impairs a minority’s ability to elect the representative of its choice.” Burton v. City of Belle Glade, 178 F.3d 1175, 1198 (11th Cir.1999). In Thornburg v. Gingles, 478 U.S. 30, 106 S.Ct. 2752, 92 L.Ed.2d 25 (1986), the Supreme Court articulated three threshold requirements for Plaintiffs seeking relief for vote dilution under § 2 of the Act: (1) the minority group must be “sufficiently large and geographically compact to constitute a majority in a single-member district”; (2) the group must be politically cohesive; and (3) the white majority must vote sufficiently as a bloc “usually to defeat the minority’s preferred candidate.” Id. at 50-51, 106 S.Ct. at 2766-67. The first of these requirements relates to whether a viable remedy exists; the second and third requirements relate to whether issues of race or color are driving election results. Nipper v. Smith, 39 F.3d 1494, 1510-11 (11th Cir.1994).
Although necessary, satisfying the three Gingles requirements is not, by itself, sufficient to establish vote dilution; § 2 further requires that the “totality of the circumstances” substantiates that a minority group possessed less relative opportunity to elect candidates of its choice. League of United Latin American Citizens v. Perry, — U.S. —, 126 S.Ct. *12612594, 2614, 165 L.Ed.2d 609 (2006). However, “it would be only the very unusual case in which the plaintiffs can establish the existence of the three Gingles factors but still have failed to establish a violation of § 2 under the totality of circumstances.” NAACP v. City of Niagara, 65 F.3d 1002, 1019 n. 21 (2d Cir.1995).
Under a totality of the circumstances analysis, we now consistently turn to the Senate Report on the 1982 amendments to the Act, which names several factors that might be relevant to a vote dilution claim. These factors include:
the history of voting-related discrimination in the State or political subdivision; the extent to which voting in the elections of the State or political subdivision is racially polarized; the extent to which the State or political subdivision has used voting practices or procedures that tend to enhance the opportunity for discrimination against the minority group, such as unusually large election districts, majority vote requirements, and prohibitions against bullet voting; the exclusion of members of the minority group from candidate slating processes; the extent to which minority group members bear the effects of past discrimination in areas such as education, employment, and health, which hinder their ability to participate effectively in the political process; the use of overt or subtle racial appeals in political campaigns; and the extent to which members of the minority group have been elected to public office in the jurisdiction.
Gingles, 478 U.S. at 44-45, 106 S.Ct. 2752 (citing S.Rep. No. 97-417 (1982), U.S.Code Cong. & Admin.News 1982, p. 177). These Senate factors are “neither comprehensive nor exclusive,” id. at 45, 106 S.Ct. 2752, but each of these factors points to election methods that may converge with social and historical factors to dilute the minority vote, Nipper, 39 F.3d at 1512.
In its dispositive order of August 27, 2004, the district court found that the Plaintiffs failed to satisfy the first Gingles requirement for a viable § 2 remedy, but satisfied the second and third Gingles requirements. Despite its belief that the Plaintiffs’ failed to surmount the threshold hurdle established by Gingles, “[f]or the sake of completeness,” the court proceeded to undertake a totality of the circumstances analysis. In so doing, the court determined that the Plaintiffs failed to meet their burden of showing that at-large elections deprived African Americans of an equal opportunity to elect candidates of their choice.
We review a district court’s findings of § 2 vote dilution for clear error. Fed.R.Civ.P. 52(a);15 Gingles, 478 U.S. at 79, 106 S.Ct. at 2781; Johnson v. Hamrick, 296 F.3d 1065, 1074 (11th Cir.2002); Negron v. City of Miami Beach, 113 F.3d 1563, 1566 (11th Cir.1997). The “clearly erroneous” standard of review appropriately affords substantial deference to the district court. “Deference is afforded the district court’s findings due to the court’s special vantage point and ability to conduct an intensely local appraisal of the design and impact of a voting system.” Hamrick, 296 F.3d at 1074 (quoting Negron, 113 F.3d at 1565). Despite the interpretative latitude granted the district court, however, review for clear error “does not inhibit *1262an appellate court’s power to correct errors of law, including those that may infect a so-called mixed finding of law and fact, or a finding of fact that is predicated on a misunderstanding of the governing rule of law.” Gingles, 478 U.S. at 79, 106 S.Ct. at 2781 (quoting Bose Corp. v. Consumers Union, 466 U.S. 485, 501, 104 S.Ct. 1949, 1960, 80 L.Ed.2d 502 (1984)).
In vote dilution cases, we require a district court to explain with particularity both the reasons for its ultimate decision as well as its subsidiary findings of fact. Solomon v. Liberty County Commissioners, 221 F.3d 1218, 1227-28 (11th Cir.2000).
III.
Having set out the legal framework for this litigation, we turn to the questions presented in this appeal. We consider first whether the district court erred in finding the Plaintiffs’ proposed plan an insufficient § 2 remedy under the first Gingles requirement for a minority group to be “sufficiently large and geographically compact to constitute a majority in a single-member district,”16 and then ask whether the court erred in concluding that the totality of the circumstances did not support a vote dilution claim.17
A.
In ruling against the Plaintiffs, the district court declared that “[njothing in the record suggests voter registrations, voting rates, or voter practices which would give real-world meaning to the statistically sufficient district. The court finds the proposed District 1 is in reality only an influence district.” We conclude that the district court erred in finding no § 2 vote dilution remedy under the first Gingles requirement where the Plaintiffs’ remedial district carried a 50.23% voting age African American majority.
Notwithstanding that blacks constitute at least a 50.23% majority in the proposed District 1, the district court found that the Plaintiffs did not meet the first condition of Gingles, that the minority group be “sufficiently large and geographically compact to constitute a majority in a single-member district.” Thornburg v. Gingles, 478 U.S. 30, 50-51, 106 S.Ct. 2752, 92 L.Ed.2d 25 (1986). The district court did not cite any support for the conclusion that a numerically majority-minority district is a mere “influence district.” Indeed, the concept of an “influence district” is directly at odds with a district in which the minority group constitutes a majority. A majority is a majority, by however small a margin, and every other court addressing the question has held a 50% numerical majority sufficient. See, e.g., Valdespino v. Alamo Heights Indep. Sch. Dist., 168 F.3d 848, 852-53 (5th Cir.1999) (utilizing a 50% bright line rule); Cousin v. Sundquist, 145 F.3d 818, 828-29 (6th Cir.1998) *1263(same); Parker v. Ohio, 263 F.Supp.2d 1100, 1104-05 (S.D.Ohio) (three-judge court) (same), aff'd mem., 540 U.S. 1013, 124 S.Ct. 574, 157 L.Ed.2d 426 (2003).
The district court dismissed the clear numerical majority by claiming that in order to be a viable majority with the ability to elect, it must be assumed that “every voting-age African American would have to be registered to vote, actually vote, and vote for the same person.” In actuality, however, the court’s conclusion is premised on several untenable assumptions; namely, that: (1) while not every black person would be registered and would vote, every white person would, in fact, be registered and would vote, and (2) all white voters would, in fact, vote for the same person and none would vote for the black minority’s candidate of choice. These assumptions are not supported by any evidence in the record.18
The proposed District 1 is composed of approximately 50.23% non-Hispanic black voters, 15.23% Hispanic voters, and approximately 33% white voters. For the district court to conclude that white voters would consistently defeat the black majority’s candidate, the district court would have to assume that a population of approximately 33% white voters would turn out in sufficient force to defeat the candidate preferred by the 50.23% of the voting population which is black. No record evidence was introduced regarding the rates of turnout of black and white voters in Glades County. Nevertheless, without any support, the district court’s analysis assumes that a greater percentage of black voters than white voters would stay home on election day.
Assuming an equal rate of turnout following the institution of the Plaintiffs’ proposed remedy — given the lack of record evidence as to black versus white turnout in Glades County — the fact that more than 50% of the voters in District 1 are black— even by a small margin — means that the district is majority-minority, particularly relative to the proposed 33% white minority. Never before in an assessment of a proposed majority-minority district has a court required the plaintiff to prove that more minorities would actually vote.19 This conjecture is clearly erroneous.
The second assumption on which the district court ruled that the numerical ma*1264jority would be unable to elect a candidate is likewise erroneous because it assumes that all white voters would vote as a bloc to defeat the black minority’s candidate of choice. The Plaintiffs introduced undisputed record evidence of consistent white crossover voting in Glades County of 19.2%. The district court stated that it found credible the Plaintiffs’ expert testimony regarding white crossover voting of 19.2%. Although this crossover percentage was not significant enough to disprove the existence of an effective white bloc vote which presently defeats the blacks’ candidate of choice (i.e., the second Gingles requirement), the crossover vote nevertheless should have been statistically significant in determining the sufficiency of the black minority’s ability to elect in District 1.20 See Sanchez v. Colorado, 97 F.3d 1303, 1319 (10th Cir.1996) (“Gingles, however, doesn’t require an absolute monolith in the Anglo ... bloc vote and recognizes the existence and role of white crossover voting. It does ask, though, whether as a practical matter, whites usually vote as a bloc to defeat the minority-preferred candidate.”). Yet in this case, the district court failed to take into account the undisputed 19.2% crossover voting rate when determining that the margin of black majority in District 1 would be too narrow to constitute an “ability to elect” district.
The Supreme Court described vote dilution in Gingles as existing where a white voting bloc defeats not only the black minority but “the combined strength of minority support plus white ‘crossover’ votes.” 478 U.S. at 56, 106 S.Ct. 2752; see also id. at 89 n. 1, 106 S.Ct. 2752 (O’Connor, J., concurring) (explaining that a minority too small to constitute a majority in a single-member district would have an interest protected by Section 2 if it can show that white support would probably enable the election of the candidates its members prefer). Again in Voinovich v. Quilter, 507 U.S. 146, 113 S.Ct. 1149, 122 L.Ed.2d 500 (1993), the Court noted that the first Gingles factor could be modified when analyzing an influence dilution claim when the plaintiffs establish a district with what is technically less than a majority of minority voters, but “the possibility of being a sufficiently large minority to elect their candidate of choice with the assistance of cross-over votes from the white majority,” id. at 158, 113 S.Ct. 1149.
Most recently, in Georgia v. Ashcroft, 539 U.S. 461, 123 S.Ct. 2498, 156 L.Ed.2d 428 (2003), the Supreme Court adopted the term “coalition districts” to refer to districts in which minority-preferred candidates can be elected with the assistance of white crossover voters. Id. at 481, 123 *1265S.Ct. 2498. The Court specifically contrasted coalition districts with influence districts, defining an influence district as one in which minority voters have some form of political “influence” notwithstanding that they are unable to elect their preferred candidates even with white crossover support. Id. The Court nevertheless held that coalition districts are not retrogressive for purposes of the Voting Rights Act, a proposition with which even the dissenting Justices agreed. Id. at 483-84, 492, 123 S.Ct. 2498 (Souter, J., dissenting). In this case, the district court erroneously ignored the 19.2% white crossover votes when it determined that the margin of black majority was insufficient to constitute a district with the ability to elect.
Given the evidence that District 1 contained a numerical black majority and that at least some white voters in District 1 would vote cohesively with black voters, the district court clearly erred. In District 1, the black majority would be able to elect the candidate of its choosing, not only because the black majority would constitute more than 50% of the population, but also because the black majority would enjoy support from white crossover votes. For these reasons, we reverse the district court’s finding that the first Gingles factor was not met and remand for findings consistent with that conclusion.
B.
As noted above, in conducting a totality of the circumstances analysis, we turn to several factors delineated in the Senate Report on the 1982 amendments to the Act. Gingles, 478 U.S. at 44-45, 106 S.Ct. 2752 (citing S.Rep. No. 97-417 (1982)). An error in the threshold analysis does not necessarily require reversal of the ultimate totality of circumstances decision. City of Niagara, 65 F.3d at 1002 (affirming in spite of clear error on a threshold factor). However, as City of Niagara Falls21 noted, “it would be only the very unusual case in which the plaintiffs can establish the existence of the three Gingles factors but still have failed to establish a violation of § 2 under the totality of circumstances. In such cases, the district court must explain with particularity why it has concluded, under the particular facts of that case, that an electoral system that routinely results in white voters voting as a bloc to defeat the candidate of choice of a politically cohesive minority group is not violative of § 2 of the Voting Rights Act.” Id. at 1019 n. 21 (quoting Jenkins v. Red Clay Consol. School. Dist., 4 F.3d 1103, 1135 (3d Cir. 1993)) (emphasis added); see also Solomon v. Liberty County, 221 F.3d 1218, 1227-28 (11th Cir.2000) (en banc) (requiring district courts to explain their totality of circumstances reasoning “with particularity”).
In the present case, the district court concluded that the totality of the circumstances failed to support the Plaintiffs’ vote dilution claim, but did not explain its totality of circumstances analysis *1266with particularity. The district court did consider each of the nine factors set forth by Gingles, 478 U.S. at 36-37, 106 S.Ct. 2752. The court found that the at-large voting system for the school board, and perhaps for the county commission, was enacted to dilute the African American vote; that voting was racially polarized; that the majority vote requirement “can operate in a manner that dilutes the minority African American vote”; and that African Americans suffered socio-economic disadvantages. The court nevertheless rejected Plaintiffs’ § 2 claim, on grounds that two African American candidates have been elected to public office; that African Americans can overcome socio-economic obstacles to political participation by attending political rallies and community meetings; that no overt or implicit racial appeals were made in campaigns; that elected officials have been responsive to the needs of the African American community; and that the policies underlying the electoral system were not tenuous. Some aspects of the court’s analysis are conclu-sory and open to serious doubt.
With regard to minority success in elections, the district court cited the elections of Charles Hall to the county commission in 1976 and of Beamon Rich to the Moore Haven city council in 1988. However, Rich was not elected under the at-large majority voting system at issue here. Instead, he ran in a city council election with no majority vote requirement and won a seat with only 16% of the vote in a city where 18% of the population was African American. The example of Rich supports the inference that African American voting power, which is concentrated in Moore Haven, has been diluted by the at-large voting system. The district court’s contrary inference is clearly erroneous. Moreover, Hall’s utility as an example of success is undermined by the fact that he was elected when the eligible African American voting population was nearly double what it is today. Coupled with racially polarized voting, the lack of minority success since Hall’s election tends to demonstrate dilution. Another probative fact — which the district court did not consider — was that only three African Americans have run for public office in Glades County history. Given a history of discrimination and socioeconomic disadvantage, the paucity of minority candidates may not be completely fortuitous. The district court’s strained analysis of minority candidacies is particularly troubling because the Supreme Court has identified the extent to which they succeed as one of the two most important factors in vote dilution analysis (the other being racially polarized voting). Gingles, 478 U.S. at 48 n. 15, 106 S.Ct. 2752.
The district court’s conclusion that African Americans faced no obstacles to political participation is also inadequately reasoned. The court found that African Americans suffered significant socio-eco-nomic disadvantages. In determining the electoral effect of that disadvantage, the court confined its analysis to transportation for political candidates. The court noted that although an African American candidate was less likely to own automobiles, she could still campaign by attending community meetings and political rallies. The court did not expressly consider other aspects of political participation, such as voter registration, voter turnabout, political connections, and participation in political parties and other political organizations. See Solomon, 221 F.3d at 1222 (considering these factors). Those factors, not automobile ownership, are ones that usually come to mind in considering obstacles to political participation. See id.; see also City of Niagara Falls, 65 F.3d at 1021 *1267(analyzing voter registration and turnout). It is difficult to reconcile the failure to consider those factors, given that the district court had implicitly assumed lower minority voter turnout to find that a 50.23% minority-majority was insufficient in the threshold inquiry. A fuller assessment of obstacles to political participation is warranted.
Elsewhere, the district court made con-elusory findings. For example, the district court stated that “the weight of the convincing evidence establishes a responsiveness to the minority community,” but did not describe any of that evidence. The district court found that the policies underlying the voting system were not tenuous, but did not discuss what those policies were. Cf. Solomon, 221 F.3d at 1234 (policy driven by citizen’s reform movement); NAACP v. Fordice, 252 F.3d 361, 372-73 (5th Cir.2001) (policy of reducing operational costs). Even if these conclusions are not clearly erroneous, the district court’s analysis should have been more developed.
When compared with the detailed district court findings affirmed in City of Niagara Falls and Solomon, it is apparent that the district court here failed to explain its reasoning with particularity to facilitate informed appellate review. See, e.g., Solomon v. Liberty County, 957 F.Supp. 1522, 1557-72 (N.D.Fla.1997). We thus remand this case for reconsideration of the totality of circumstances in light of the district court’s error in the threshold Gingles inquiry. See Houston v. Lafayette County, 56 F.3d 606, 613 (5th Cir.1995) (remanding after finding error in threshold inquiry); Jenkins, 4 F.3d at 1135 (same).
IV.
The district court clearly erred in finding that District One of the Plaintiffs’ illustrative plan constitutes an influence district. Thus, the court committed error in concluding that the Plaintiffs failed to establish a § 2 remedy. Furthermore, the court failed to explain with sufficient particularity that the totality of the circumstances weakens the Plaintiffs’ vote dilution claim. In making those determinations, the court did not properly apply the relevant legal principles and grounded its findings in inaccurate perceptions of the law. We therefore reverse the district court’s holding that the Plaintiffs’ proposed remedial plan is insufficient under the first prong of the Gingles test and remand to the district court for reconsideration of the totality of the circumstances test.
REVERSED AND REMANDED.
. The Hispanic population in Glades County appears to be on the rise, significantly outpacing the modest growth of the African American population. In ten years (1990-2000), the African American voting age population rose from 9.6% to 10.1%, while the Hispanic voting age population eclipsed the African American voting age population, increasing from 6.7% to 12.7%.
. The at-large voting scheme provides that all voters in the county may cast ballots for any of the seats on the County Commission or School Board (and for at most one candidate per district).
. In the same year, African American candidate Beamon Rich won election to the City Council for the City of Moore Haven. Rich placed third in a seven-candidate race with 16.2% of the vote; the three candidates who receive the most votes are elected to the City Council.
. These claims for relief are asserted together rather than in separate counts.
. The complaint alleges that the at-large election schemes for the County Commission and School Board, "including the use of majority vote, residency districts, and staggered terms, with no provision for limited or cumulative voting or other non-majoritarian election procedures, results in the denial or abridgement of the [plaintiffs’ right] to vote on account of race or color.”
. The Equal Protection Clause of the Fourteenth Amendment states that no State shall "deny to any person within its jurisdiction the equal protection of the laws.” U.S. Const. amend. XIV, § 2.
. The Fifteenth Amendment states: “The right of citizens of the United States to vote shall not be denied or abridged by the United States or by any State on account of race, color, or previous condition of servitude.” U.S. Const. amend. XV, § 1.
. The record, suggests that the Plaintiffs proposed only one illustrative plan at trial.
. Cooper did not testify; instead, he proposed a report which was introduced into evidence without objection. The report contained the Plaintiffs' illustrative plan based on 2000 census figures from the Census of Population and Housing PL 94-171. The parties stipulated that the districts drawn in the plan are compact, contiguous, and otherwise follow the requisite redistricling criteria.
. Cole analyzed seven election contests involving African American and white candidates that took place from 1976 through 1998: the 1976 Democratic Primary for County Commission, the 1984 Presidential Preference Primary, the 1988 Presidential Preference Primary, the 1990 Democratic Primary for Secretary of State, the 1994 Democratic Primary for State Commissioner of Education, the 1994 General Election for State Commissioner of Education, and the 1998 Democratic Primary for School Board District Four.
.Cooper’s statistics defined African Americans as "non-Hispanic blacks.”
. The district court found no evidence of either a discriminatory purpose or discriminatory results in the enactment or operation of election procedures in Glades County, and therefore, denied the Plaintiffs’ claim under the Fourteenth Amendment. Furthermore, the district court noted that in the Eleventh Circuit, vote dilution is not a cognizable claim under the Fifteenth Amendment. See Osburn v. Cox, 369 F.3d 1283, 1288 (11th Cir.2004) (“The Supreme Court has recognized that the Fifteenth Amendment protects the right to register and to vote, but it has never held or even suggested that vote dilution violates the Fifteenth Amendment.”) (quoting Reno v. Bossier Parish Sch. Bd., 528 U.S. 320, 334 n. 3, 120 S.Ct. 866, 875 n. 3, 145 L.Ed.2d 845 (2000)).
. The regression analysis that accounted for Hispanic voters concerned only the 1998 Democratic Primary for School Board District Four, the election in which plaintiff Billie Thompson participated.
. “Findings of fact shall not be set aside unless clearly erroneous, and due regard shall be given to the opportunity of the trial court to judge of the credibility of the witnesses.’’ Fed.R.Civ.P. 52(a).
. We need not address the district court’s discussion of the second and third Gingles requirements, because the court found that the Plaintiffs satisfied both, and the Plaintiffs raise no question on appeal regarding either requirement.
. Plaintiffs also contend that the district court clearly erred in finding that the 8.6% standard deviation within the illustrative plan constituted more than de minimis deviation under the one-person, one-vote principle of the Fourteenth Amendment, and therefore, the plan failed to satisfy the first Gingles requirement for a viable § 2 remedy. The standard deviation finding is not dispositive for the district court's conclusion that the Plaintiffs failed to establish an appropriate remedy, however, and we therefore decline to address the issue here.
. The district court also erred in its determination that District 1 would contain only a 50.23% majority because it counted as "black” only non-Hispanic blacks, omitting persons who self-identified on the 2000 Census as both black and Hispanic. In reality, then, there may be significantly more than a 50.23% black majority in District 1. In Georgia v. Ashcroft, 539 U.S. 461, 123 S.Ct. 2498, 156 L.Ed.2d 428 (2003), the Supreme Court explained that in cases where the dilution of black votes is at issue, anyone who checked off black, even if they checked off some other category also, should be counted as black. Id. at 473 n. 1, 123 S.Ct. 2498. The district court’s legal conclusion that District 1 was majority-minority by too small a margin rested on clearly erroneous factual assumptions about who constituted the relevant minority.
. The Supreme Court’s recent analysis in Perry that a district with a bare Latino majority failed to satisfy Gingles is inapplicable because the Supreme Court's conclusion that many of the Latinos were not citizens eligible to vote is different from the district court’s conclusion here that eligible voters might not turn out. See 126 S.Ct. at 2616. There is a crucial distinction between a district with a majority of eligible minority voters and a district that is only majority minority because non-citizens are included in the count of the minority population. There is no dispute in this case that District 1 constitutes the former, not the latter.
. The third Gingles prong does not preclude courts from considering vote dilution claims for coalition districts. The Court in Gingles specifically considered white crossover voting and found prong three still satisfied as long as "a white bloc vote ... normally will defeat the combined strength of minority support plus white 'crossover' votes." 478 U.S. at 56, 106 S.Ct. 2752. In other words, it is possible for there to be white crossover voting that is insufficiently substantial to defeat the Plaintiffs' assertion of white bloc voting under the second Gingles requirement but that is nevertheless significant enough to establish an "ability to elect” district under the first Gin-gles requirement. Cf. Cane v. Worcester County, 35 F.3d 921, 926 (4th Cir.1994) (finding average 19% white crossover vote insufficient to salvage at-large, one-on-one election scheme under the facts of that case), cert. denied, 513 U.S. 1148, 115 S.Ct. 1097, 130 L.Ed.2d 1065 (1995). "If blacks constitute just under fifty percent of the population and a small fraction of white voters are willing to cross over, voting is still clearly racially polarized, yet black voters can prevail in a coalition district.” Note, The Implications of Coalitional and Influence Districts for Vote Dilution Litigation, 117 Harv. L.Rev. 2598, 2605 (2004).
. In City of Niagara Falls, the district court considered a challenge to an at-large voting system for the seven-member city council. The district court found no history of official discrimination, no unusual voting practices or procedures, no racial appeals in campaigning, some success for minority candidates in elections, and responsiveness on the part of public officials to minority needs, and no tenuous policies underlaying electoral practices. These factors outweighed racial polarization, the socioeconomic disadvantages of the African American community, and the possible dilutive effect of staggered elections. The court affirmed only after closely examining the district court's detailed totality of the circumstances analysis. 65 F.3d at 1019-24.